The long term goal of the studies in this proposal is to develop methods for enriching endoderm-derived pancreatic beta cells and their progenitors from murine embryonic stem (ES) cells, with the intent of potentially using these cells for clinical applications such as islet transplantation for type I diabetics. The endocrine pancreas is derived from definitive endoderm of the embryo. The Sry-related HMG box containing transcription factor Sox17, an endoderm marker, appears to be important for the development of endoderm and the specification of pancreas. Of particular relevance to this grant application, we have developed novel culture methods that permit the commitment of ES cells to the early endocrine pancreas in vitro (Stem Cells, 22:1205). This ES-derived population expresses Sox17, pdx-1 (a pancreatic marker), insulin I and insulin II. Addition of beta cell specification and differentiation factors in culture increases insulin-expressing cells to 2.7% of the total population. However, the direct lineage relationship between the Sox17-expressing endoderm progenitors and the endocrine pancreas has not been established, and the conversion efficiency of pancreatic insulin-expressing cells remains low in our current culture system. We propose in this pilot/feasibility application to generate novel tools that could be used to define endodermal commitment to the endocrine beta cells in vitro. In particular, we will test the hypothesis that efficient specification of ES cells to Sox17-expressing endoderm progenitors is required to generate pancreatic endocrine stem cells and their progeny. The studies outlined in this proposal will: 1) generate double transgenic AinV ES lines that contain endogenous Sox17-driven enhanced green fluorescent protein (EGFP) and insulin l-driven red fluorescent protein (RFP) reporter genes to allow detection and purification of endoderm progenitors, and easy identification and characterization of committed beta cells in culture; 2) sort and enrich Sox17-EGFP+ cells to allow efficient induction of endoderm-derived pancreas. We will test whether the sorted Sox17-EGFP+, but not the EGFP-, cells can give rise to pancreatic insulin I-RFP+ cells that co-express pdx-1 and insulin II. This in vitro system will be a novel tool to study mechanisms that control specification of beta cells, and lead to the development of unique methods for producing endocrine stem cells and their progeny for potential therapeutic applications. [unreadable] [unreadable] [unreadable]